Task Allocation in Clusters of Cognitive Nodes: A Remuneration-Aided Approach

Abstract

In this work, we propose a remuneration-aided Game theoretical solution for task allocation in cognitive radio (CR) enabled Internet of things (IoT) scenarios, where cognitive nodes (CNs) in close proximity and with similar sensing capabilities are clustered around a cluster head (CH). We consider a framework in which task allocation in the system is driven by CNs with spectrum sensing capabilities. In the proposed approach, the CH assigns a remuneration to CNs for their contribution in spectrum sensing prior to initiating the task allocation procedure. Such remunerations can be used by CNs in proposing the bids to win the task in the Game. Hence a non-cooperative Game approach modelled as an auction process is proposed. We show that the proposed framework is able to exploit cognitive behaviour efficiently in conditions suitable for cognitive radios (low spectrum occupancy), and under the same conditions the overall system utility increases by 29% w.r.t the case when licensed users (LUs) occupy the band 70% of the time. Additionally, the framework allows the system to reap benefits of energy efficiency while experimenting cognitivity.